Simple uptime monitor for small remote network

Question

Our company has about 15 small stores with two Windows 7 PCs (and two VoIP phones) at each location behind a SonicWall SOHO. We use various business grade cable and fiber ISPs at each location. Occasionally our store managers complain about the "internet not working". I would like to install a small systray app on one PC at each location that simply verifies the computer is online (say once every 5 minutes) and sends the log to a server. The server would then show a chart of each location and their corresponding historic uptime. Ideally we would get an alert (via Slack would be ideal) if a location is offline for more than 30 minutes so we can send out tech out.

This is simple enough for us to get an idea of the problem but ideally we would want to know if the ISP is having connection problems, the SonicWall SOHO is having problems or the PC is down.

We have static IPs at most locations and thus I could ping the SonicWalls but I was hoping for an agent app.

I've read about nagios, what's up gold, spiceworks, pagerduty, uptime robot and others but not sure if any of those are simple enough to do what I mentioned above.

Update: I'm looking for a solution that doesn't require VPN tunnels and ideally no server that I also need to maintain. Ideally it would be a SaaS product like New Relic but for PCs (not web apps) and free for less than 50 computers.

Update 2: I'm also wondering if there is a solution using built-in Windows tools and Azure AD.

Answer 1

My current solution is to use a tool like Pingdom or Uptime Robot (free for less than 50 monitors and alerts to Slack) to ping each SonicWall every 5 min.

Since each location has a static IP, all I had to do was allows Ping responses in the firewall setup of each location and create little monitors.

Thus, I will not be verifying the PCs behind the firewall are working, but I will be monitoring that the Internet connection and firewall are both functional.

Nothing fancy, but ticks the boxes simply and free.

Answer 2

What I understand, from your question, is that you need a MONITORING INFRASTRUCTURE. Even if you're describing it with different wording, it's clear that you need something that:

• periodically check if your 15 remote sites are "up-and-running";
• should some nodes be off-line, raise some alarms (send e-mail, SMS, etc.);
• keep historical tracks of status/events.

Well.... The above IS a monitoring infrastructure.

You lasted with: "not sure if any of those are simple enough to do what I mentioned above" and there, the word "simple" may represent an issue: I'm definitely convinced that putting in place a very-basic-monitoring-platform is not exactly a "simple" task but... nevertheless, if you feel the need of such a platform, than you are mostly forced (for your own benefit) to scale-up, investing time and resources in implementing such a platform.

So, back to your question, here is my proposal: connecting all the 15 remote sites via VPN to the headquarter, where a monitoring-box will periodically check the health of your network/services/remote_sites. The VPN server and the monitoring BOX could live in the same Linux box and the solution will be 100% open-source based.

In detail:

1. Install a linux box (even a VM is OK) and install/configure an OpenVPN server (with certificate authentication);

2. Install the OpenVPN client on each of the 15 win7 box. Let the client start at boot, so every time the Win7 box will restart, it will promptly connect to the OpenVPN server;

3. properly configure the windows firewall, so to accept any kind of traffic coming along the VPN;

When the three steps above will be completed, you will have full-access to every Win7 box, from your headquarter (via the linux, openvpn server).

With such a network infrastructure in place, you're ready to deploy the monitoring platform.

I generally avoid suggesting products and technologies but... to properly answer your question, it's really needed to mention some products.

Two products, in my opinions, should fit well in your context, granting you the flexibility that... now you don't really need, but I expect you'll need in upcoming months: NAGIOS and ZABBIX. They're really different products, with different approaches to monitoring tasks.

Of such two products, considering that your main focus is the "Win7 box" (and not the "remote network and resources"), I suggest going with ZABBIX. So:

4. install Zabbix on the linux/openvpn box. Chances are very high that it's already included in the software repository, so the installation is really simple;

5. deploy the Zabbix-agent on every Win7 box, being careful in configuring them so to answers requests coming from the zabbix-server;

and, lastly:

6. spend some time (several days, at least) configuring zabbix so to:
   • access the 15 windows box (zabbix "hosts") via zabbix agent;
   • retrieving the status of entities you want to monitor (CPU, memory, I/O, network status, etc. - zabbix "items");
   • defining "alarm" conditions (zabbix "triggers")

When all of this will be in place, you will be able to know the status of your remote sites, very easily, from your headquarter (or even from your smartphone, as OpenVPN client is available also for Android and iOS)

In the end:

• is the above "easy/simple"? No. Not really. But it's also not really complex;

• is the above "worth-the-effort"? Yes. Definitely. You'll need some time to put everything in place but... when everything will be ready, you'll start thinking how have you survived, before, without such an infrastructure;

• is the above "the best monitoring infrastructure of the world"? Probably not. But it's one that fits very well with your environments (In my humble opinion);

• what if you'll have problem in installing/configuring it? The good news is that all the software components I suggested (Linux, OpenVPN, Zabbix) are well-known open-source projects. Even here, in the great ServerFault community, they're well represented [see here, here, and here]. Also, the web is plenty of really good documentation about all of them.

Good luck! And welcome in the (nice, BTW) system/network administrator world :-)