OPNsense Forum
English Forums => General Discussion => Topic started by: rhambus on January 04, 2021, 04:42:57 pm
-
Hey all,
New to OPNsense here, but my wife (and kids) are complaining that the internet keeps going down on them (they complained about this before I was on OPNsense too so it's not that). I suspect it's the WAN that is going up and down (it's AT&T gigapower) and feel like there has to be some solution to simply see a graph or table or something of uptime. What I am picturing is, say, OPNsense pings 8.8.8.8 every 30 seconds and keeps track of missed pings. I searched here a bit, and while some recommended Monit, that seems to be about emailing when something happens, not simply logging it. I also tried to enable Netdata, but that service won't even start. Surely I am not the only person who wants to be able to do this - any recommendations? I know there are monitor programs that you can run in the cloud that ping your firewall, but I'd like the firewall to ping something else, for security reasons if nothing else. I'd really appreciate any advice or leads!
Thanks!
-
well i read that post and thought to myself "no way that's not in there". could've sworn i've already seen and used it. and lo and behold, the feature seems to have vanished?
it's even in the docs (https://docs.opnsense.org/manual/systemhealth.html#data-collectors) but nowhere to be found on my 20.7.7_1 installation (including today's hotfixes).
is this an (un-)intentional side effect of introducing the new traffic graphs?
/e:
After a bit of fiddling i found out what the issue was. The WAN interfaces on my test VMs are all just set to DHCP. This automatically creates gateways but gateway monitoring is disabled for them, for reasons i don't yet fully understand. Might be technical limitations due to how gateways are handled.
Anyway, if you simply edit the gateway and untick the "Disable gateway monitoring" checkbox and save, the "Quality" tab will appear. In order to determine the source of error for your WAN problems, you might want to play around with different monitor IPs here. I'd start with the first hop address of your ISP. If nothing suspicious is going on there, then move to something like a well known DNS resolver. Always use the graphs to determine whether there's an upstream failure or outage.
-
Gateway monitoring is off by default stemming way back to issues with the poor apinger service performance.
Cheers,
Franco
-
Not sure if this will work for you, but apart from Gateway monitoring I also use a couple of free services that monitor my gateway or servers from the internet inwards, in the image shown below it's monitoring my IPv6. Not sure if you can do anything with your AT&T hardware, but I also monitor my modem connection too. Check around and see if anyone is doing similar for your AT&T hardware.
(https://www.thinkbroadband.com/broadband/monitoring/quality/share/thumb/7b35a2f1cf51230cf7a649c6eea17b4fdb223311-04-01-2021.png) (https://www.thinkbroadband.com/broadband/monitoring/quality/share/7b35a2f1cf51230cf7a649c6eea17b4fdb223311-04-01-2021)
It's a UK based site but there must be something similar where you are. You can easily see where we are busy during the working day.:)
-
Gateway monitoring is off by default stemming way back to issues with the poor apinger service performance.
Cheers,
Franco
Thanks to both of you. I guess I should keep off the gateway monitoring, then?
-
Not sure if this will work for you, but apart from Gateway monitoring I also use a couple of free services that monitor my gateway or servers from the internet inwards, in the image shown below it's monitoring my IPv6. Not sure if you can do anything with your AT&T hardware, but I also monitor my modem connection too. Check around and see if anyone is doing similar for your AT&T hardware.
(https://www.thinkbroadband.com/broadband/monitoring/quality/share/thumb/7b35a2f1cf51230cf7a649c6eea17b4fdb223311-04-01-2021.png) (https://www.thinkbroadband.com/broadband/monitoring/quality/share/7b35a2f1cf51230cf7a649c6eea17b4fdb223311-04-01-2021)
It's a UK based site but there must be something similar where you are. You can easily see where we are busy during the working day.:)
Yes, I was hoping for a solution going the other way, but I may have to resort to something like this. If anyone else has suggestions for internal solutions, though, please keep them coming!
-
I Should add I also monitor my modem using a separate client on one of my servers, I also use Monit on the gateway dpingers to watch that way too. Any which way I got a notification of a WAN failure and can identify the reason why quickly.... though it is a very rare occurrence.
-
Gateway monitoring is off by default stemming way back to issues with the poor apinger service performance.
Cheers,
Franco
Thanks to both of you. I guess I should keep off the gateway monitoring, then?
why? apinger has been replaced with dpinger and performance should be fine now. i guess the default has just been kept as-is to not introduce unexpected behaviour to existing systems.
-
It seems safer to leave it disabled by default. I cannot remember related discussions that this default broke something. Cannot assume this for dpinger now the other way around, but then again it's quickly enabled if required.
Cheers,
Franco
-
Graphing of network stuff on OPNSense? Wasn't ntopng born to do exactly that? :)