Installation corruption

General
21

My sense is that both times it happened after I made some change in the flow and attempted to deploy - I can see how it could have been some feedback loop or other condition etc. What I don't understand is why there seemed to be "damage" that remained after server reboot.

So what does that "node-red restart" actually initiate?

22

If you manage to do something to flows that cause an issue (like a loop) then restarting the server simply restarts node-red and the bad flow.

Instead, you should (from a console) stop node-red then launch it with --safe flag to start node-red without flows running. Then you can fix the flow & deploy.

How you stop node-red depends on how it is running.
For example node-red-stop is a helper command installed by the node-red installer script (if you used it).
For you, it may be sudo service node-red stop or sudo systemctl stop node-red

23

Thanks, I think in my case its indeed sudo service node-red stop or sudo systemctl stop node-red

24

Note that if something is hanging or looping in node-red it may take some time to stop, as something may need to timeout first. Therefore you should check that it is indeed stopped by using, in your case
sudo systemctl status node-red
to check.
Just to avoid confusion for those reading this I must point out that for the usual Debian/Ubuntu/Pi install the service is nodered, not node-red.

After having checked that it is stopped then, if you want to prove that it isn't something in the flows then rename the .node-red folder and start node-red again. It should start with empty flows file. If it still mis-behaves then it is something more significant.

It may be a problem with your flows that is only triggered by some external state, which explains why it seems to get locked in the mis-behaving condition.

25

Thank you.

26

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.