Phone registrations
Which handsets are signed in, and which have quietly dropped off.
Intelligence
A live view of registrations, trunks, active calls and server load, with alerts that fire before your staff notice anything.
A desk phone loses its registration overnight and nobody notices until someone tries to use it. A trunk drops one of its two routes, so calls still work but there is no longer anything to fall back on. Disk space creeps up for a month and then fills on a Friday afternoon.
None of these are dramatic while they are happening. All of them are visible, if something is looking. That is what monitoring is: watching the things that go wrong slowly, so they can be fixed before they go wrong suddenly.
A wallboard you can leave on a screen: what is registered, which trunks are up, how many calls are in progress, and how hard the server is working.
Continuously, on every system we manage.
Which handsets are signed in, and which have quietly dropped off.
Whether each route to the outside world is up, and whether failover is still available.
Calls in progress now, and the pattern over the last hours and days.
Processor, memory and disk, so a slow problem is caught while it is still slow.
Signs of a network struggling: jitter, packet loss and one-way audio.
Repeated failed registrations and the sort of probing that comes before toll fraud.
Alerts
Every alert has a threshold you can set, and a route you choose: email, a message to a team, or straight into our support queue. The aim is a small number of alerts that all mean something, rather than a stream nobody reads.
Examples of what can raise an alert
Managed customers
For the systems we manage, these alerts come to us first. Often the first you hear of a problem is a message saying it has been dealt with. If we need something from you, such as access to a cupboard or a power cycle at your end, we will ask.
For most organizations the honest answer is nothing. We can change that.