Wish? - Notify after outage

[PeterHR]

Can we have options for 3 hours and 1 hour please

The present 6 hours is long enough that I miss miss the email or the opportunity to restart 'till the following day. sadly though I might notice the stats page not being updated, that is often due to server lag.

[csweeney05]

@PeterHR

Can we have options for 3 hours and 1 hour please

The present 6 hours is long enough that I miss miss the email or the opportunity to restart 'till the following day. sadly though I might notice the stats page not being updated, that is often due to server lag.

I second that, 6 hours is too long to go and not know the system isn’t reporting.

[tjowen]

PeterHR,

We need an app that would allow one RPi to ‘keep watch’ on another RPi and report if anything unusual is going on. Each RPi would report to the other on a regular basis, say every 10 minutes. If one RPi reports that dump1090 has stopped or fails to report at all the monitoring RPi would blast off an email to the person-in-charge.

A person wouldn’t be required to own two RPi’s to make this work. You could simply pair-up with another single RPi owner and monitor each others systems.

I don’t know if such an app exists, but it would be useful.
…Tom

[LudgerBoergerding]

Yesterday I did configure the hardware watchdog on my raspberry pi 2. Sometimes my raspberry have an unusual state and I cannot login or anything else.
I hope it will work. You also can configure the watchdog daemon to check log files if they are written from time to time.

[dbaker]

It’s on the todo list, but first we need to extent the outage problem to be able to better detect if the problem is on FA’s side. We don’t want to annoy a bunch of people at the 1hr mark because we’re in the middle of an overnight maintenance window, for example.

[csweeney05]

@tjowen

PeterHR,

We need an app that would allow one RPi to ‘keep watch’ on another RPi and report if anything unusual is going on. Each RPi would report to the other on a regular basis, say every 10 minutes. If one RPi reports that dump1090 has stopped or fails to report at all the monitoring RPi would blast off an email to the person-in-charge.

A person wouldn’t be required to own two RPi’s to make this work. You could simply pair-up with another single RPi owner and monitor each others systems.

I don’t know if such an app exists, but it would be useful.
…Tom

It’s about more then Pi users though, we all crash and have issues at times with all hardware platforms.

[csweeney05]

@dbaker

It’s on the todo list, but first we need to extent the outage problem to be able to better detect if the problem is on FA’s side. We don’t want to annoy a bunch of people at the 1hr mark because we’re in the middle of an overnight maintenance window, for example.

Just turn the monitoring off when you are in a maintenance window?

[dbaker]

@csweeney05

@dbaker

It’s on the todo list, but first we need to extent the outage problem to be able to better detect if the problem is on FA’s side. We don’t want to annoy a bunch of people at the 1hr mark because we’re in the middle of an overnight maintenance window, for example.

Just turn the monitoring off when you are in a maintenance window?

Trying to avoid human factors.

[brianroyal]

I concur. Please add a shorter interval than six hours.
Because users of the PiAware can select longer intervals, opting in to more timely alerts is not a problem. Please please please please please please please please please please please please please please please please please. Please.

[tdrane]

@LudgerBoergerding

Yesterday I did configure the hardware watchdog on my raspberry pi 2. Sometimes my raspberry have an unusual state and I cannot login or anything else.
I hope it will work. You also can configure the watchdog daemon to check log files if they are written from time to time.

First, I apologize for responding to such an old comment but the thread as been awakened by another user so anything is fair game

I’ve had this issue from time to time, that of not being able to ping or SSH to the pi yet my stats page continues to indicate that it has received data in the previous minute.

Those times I’m not in house, I find that it “heals” over time. Normally though I just restart the Pi with a power cycle.

I believe the problem has something to do with waking up a computer which has gone to sleep while Firefox is on the local Dump1090 web page. I saw this happen when I was running the original Piaware supplied Dump1090 and more recently with Dump1090-Mutability and Lighttpd. FWIW, my Pi is mounted outside and connected to my network with a wireless adapter. Unfortunately (or perhaps, fortunately) the situation does not occur often enough so as to get a handle on what might be the actual cause.

[LudgerBoergerding]

@tdrane

@LudgerBoergerding

Yesterday I did configure the hardware watchdog on my raspberry pi 2. Sometimes my raspberry have an unusual state and I cannot login or anything else.
I hope it will work. You also can configure the watchdog daemon to check log files if they are written from time to time.

First, I apologize for responding to such an old comment but the thread as been awakened by another user so anything is fair game

I’ve had this issue from time to time, that of not being able to ping or SSH to the pi yet my stats page continues to indicate that it has received data in the previous minute.

Those times I’m not in house, I find that it “heals” over time. Normally though I just restart the Pi with a power cycle.

I believe the problem has something to do with waking up a computer which has gone to sleep while Firefox is on the local Dump1090 web page. I saw this happen when I was running the original Piaware supplied Dump1090 and more recently with Dump1090-Mutability and Lighttpd. FWIW, my Pi is mounted outside and connected to my network with a wireless adapter. Unfortunately (or perhaps, fortunately) the situation does not occur often enough so as to get a handle on what might be the actual cause.

Maybe energy saving function of your wireless adapter… There is a thread here in the forum about this.

[tdrane]

Maybe energy saving function of your wireless adapter… There is a thread here in the forum about this.

Thanks for the tip. I had disabled PS for the adapter in my initial troubleshooting of the problem. I am now keeping a scrupulously accurate log (well, as accurate as one can be at 5 a.m. before coffee.) I’ll report back.

[r2zac]

I join the request. Let the feeders will chance to choose a smaller period of time. At least this opportunity will be seized by those who have the ability to check your radar remotely. I agree not to complain about the warning letter, if the radar will work properly

[dbaker]

We will do this as soon as we can better detect if the problem is on our side.

[n5psp]

Me three, me four, me 65,535.

Waking up this 6 year old thread because it is still an ongoing issue here in 2021.

How about an SMS message option that can be set to a fairly hair trigger, say, as little as 10 minutes?

Another option I have thought about as a workaround, is a variant of this:

How to Turn Your Raspberry Pi Into a Network Monitoring Tool

Network monitoring with Raspberry PI and NEMS (Nagios)

[foxhunter]

Check network every five minutes, reboots if not available:

wiedehopf/adsb-scripts/blob/master/pingfail.sh

[keithma]

I have PRTG running and that pings me if my tracking Pi goes offline for more than ten minutes.

[foxhunter]

For alerting, reporting and processing different flavors of services i recommend Monit.

https://mmonit.com/monit/

Have it running on all my devices monitoring different services

[n5psp]

Looks like it’s started as a service on boot-up.

wiedehopf/adsb-scripts/blob/master/pingfail.service

I’d modify the script to also ping 8.8.8.8 and 8.8.4.4, and something local in the network, such as a printer that is always on, to prevent endless loop rebooting when the ISP is “having technical difficulties”.

Only when the Pi is also not finding the local network, then do a reboot.

The same script and service could be placed in a Linux based router, assuming the router doesn’t lock up in a manner that prevents other tasks from executing.

This is also a potential problem with the reboot script in the PiAware - if it is completely locked up, it needs an external supervisory system that can perform a hardware reset to force a reboot, much like a long hold power button kind of reboot.

This is where a separate Raspberry Pi fitted with a relay hat that switches the power to the bridge router and the PiAware, separately, can do a power-cycle reboot of either device that it determines is not responding, modified by whether or not other “always on” devices in that branch of the network, such as printers, can be pinged.

[foxhunter]

Much effort for just monitoring a few aircraft as a hobby

The script on my Raspberry is pinging the router and an external adress.