Have a doozy…
I have a client.
They have a very basic setup. 10 users, a server, a few printers all connected to an unmanaged switch. The switch is then connected to a dsl carrier modem/router (I know…don’t get me started)
About 2 months ago, they started to experience DSL problems. It had been rock solid for years. I got the DSL provider to check the line. They did (as far as the front door) and all checked out. I then bridged the router into a modem and put a EdgeX router behind it with a basic setup. Again, problems still occured but not as frequently.
I noticed that the LEDs on the modem were very faint so I swapped it out with a similar modem. Problem appear to stabilise for a bit.
I then put in a pfsense box about a week ago with a 4g (LTE) failover modem connected to one of the ports. That appeared to work for a couple of days and then during the week both Internet connections dropped. I ended up going on site, pulling the pfsense box and putting the 4G modem back into router mode and ran the network off that. It was slow but it did provide internet access.
Over the weekend, I went back and double checked everything. I had brought home the pfsense to check the config, just to make sure I hadn’t setup anything wrong. I hadn’t but I did a complete reinstall. While onsite,I noticed that the WiFi was down. I checked the access point (which is relatively) new and and wasn’t able to access it by way of the web configuration page. I tried a hard reset but no luck. The access point appears to be borked.
This morning, I log into the network remotely, I see the WAN is down but the failover 4G modem (back in bridge mode) is up.
But here’s the really interesting thing. Anyone who uses screenconnect for remote access knows that it connects back to a connectwise instance. All the machines have screenconnect on them. I was able check the connection ‘timeline’ of each of the screenconnect clients. This would show if there was an break in the connection between the client and the connectwise server.
During last night.
4 of the client computers were relatively stable. The connection to the screenconnect server had a few small disconnects, we’re talking maybe 30 secs. This proves there was an internet connection.
5 of the others experienced drops between 7 and 40 minutes. They are all in different parts of the building
So the internet is a problem for sure but there’s more going on here.
I’m thinking it might be power surges. Based on the 2 other devices (modem and wireless access point) proving troublesome and needing to be replaced.
One other point that I haven’t mentioned but might be pertinent is that there was renovation work being carried out in an adjoining unit during this period. They have stopped about 2 weeks ago but I’m just wondering if there’s anything they could have done to somehow alter the power.
I hope my explanation is somewhat coherent. It would be great to hear any ideas or theories. This has worn me down.
Thanks.
Niall
ps. does anyone know or have experience of using power monitoring devices?