Sudden severe failure of entire flow on machine

Hey folks.

Bad new where I am: One of my machines now won't run the old NR flow it did a couple of days ago.

Situation - bigger picture:

RasPi 3B+ running Buster.
It is my NTP server, MQTT broker and a WAP.

Weirdly, it loads, starts and runs, but keeps restarting NR.
(NOT rebooting!)

I tweaked something (wrongly) with the NTP stuff and realised it wasn't needed.

Alas between doing it and realising it: all hell broke loose.

Where the machine usually runs at 12% CPU load, it maxed out at 50, 60,70% CPU load.

I was looking at the dashboad (old) from the remote machine and it kept showing me "connection lost" then would come back soon after.

This was unusable as I also lost my local DHCP server and so remote access was kaput.

Powered it down, unplugged everything and directly connected to it with keyboard, mouse and monitor.

Booted. CPU load way up.
Loaded .... chrome (Can't remember if it was chrome, or opera)... When I clicked on the "stock" browser (I think it is chrome.... but blue circle icon) It said something about Firefox, but I didn't make sense of it so just continued.

Running the browser on that machine is kinda pushing it but I have done it.

Loaded the dashboard - slowly - and the edit screen.
After a few seconds the browser ...... "connection lost" and came back a short time later.
So this is proof the machine didn't reboot.

Here are some (maybe) useful screen shots.

This is one where it restarted

I was sure I caught more, but I am running around like a headless chicken just now.

Other things of interest:

It was running and I kept getting voice messages and I would clear them.
Then soon after they would happen again.
I was trying to turn off all the logging and as I was doing it NR would restart.

I'll work on getting a log to include/post but I didn't see anything worrying.
(But what do I know/) :wink:

Suspecting a problem I deleted all the flow stuff and put an older flow in and ran it.
No luck.
Same outcome.
HIGH CPU load, NR restarts about every 1 minute.
Which I think is shown in the last screen shot.

There may be more than 1 issue, but it is restarting NR due to a problem with the tts-ultimate node.
Try disabling that node in your flows, and put a debug BEFORE it so you can see what is being sent to it.

1 Like

Just quickly.... ok, I could look and check myself...

node-red
node-red-start

Which is which?

One only runs until you ^c and the other way it keeps running.

So to do what you suggested:

I'd node-red --safe and disable the node then DEPLOY and see what happens - yes?