Ok i will try it…
Pierre RUWET
Did that help to determine what was going on?
Perhaps the timezone was not set correctly, for example.
Hello Colin,
Yesterday, i spent time to check your proposal.
I added the debug node sending 'Restart' to console.log on the same inject node that send the Restart message to my telegram account.
Then i changed my setting.js log config from "fatal" to "trace" to have maximum log content and restarted Nodered by resarting the nodered container on docker.
I could see clearly the Restart recodrede by debug but also the system message when nodered starts like
pi@raspberrypi:~ $ grep 'Node-' nodered_log3.txt
2022-10-07T18:10:43.524886291Z Welcome to Node-RED
2022-10-07T18:10:43.526682076Z 7 Oct 20:10:43 - [info] Node-RED version: v1.2.6
then I tried to check is there was a real Nodered restart in the case i observed on the 30th of september where i had the chance to save log to files after telegram indicated a Nodered restart but the grep did not find anything.
So i finally aggree with you that Node-red is not restarting by itself in my case.
What i have is that some inject nodes that run only once after nodered start are sending message even nodered is not really restarted. Is it possible to edit the title of this thread to make it match the case really observed?
Now i just have to wait for the next time it appear to check exactly which initial process are really triggered in this case (for example, at node rede restart, my OPC UA node send Browse requetst to OPC UA server to get the list of item to subscribe. This appear clearly in the log file.
I am not sure you have proved that yet, though it may be true. I think you will need to wait for it to happen again and see if the debug message after the inject node appears in the log.
Yes, click on the pen icon next to the title.
Hello I come back for some follow up.
I focus last weeks on the migration to nodered 3.
As I am running on docker, I created a new container with a separate instance running the new nodered.
This new instance seems working better. When migrating my flows, I reorganized them to reduce the number of OPC UA subscription that seemed to generate a lot of nodered errors.
My new nodered was running well for weeks but I noticed a restart last week, this time I can confirm that the nodered completely crashed because the container was also restarted.
You need to look at the node red log at the crash to see why it crashed.
It seems unfortunately’ I should have checked immediately after the reboot …OPC UA client is continuously filling the log with not useful warning.
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.
