[info] Node-RED version: v0.18.4
20 Jun 09:31:46 - [info] Node.js version: v6.12.3
20 Jun 09:31:46 - [info] Linux 4.4.105-cip15 ia32 LE
20 Jun 09:32:00 - [info] Loading palette nodes
20 Jun 09:33:03 - [info] Dashboard version 2.9.1 started at /edge
i have noticed that node-red stops working after some hours, i set up in the settings the trace option for the logging but there is nothing significant there that explains me this behaviour.
What should i do to solve this issuue??
every suggestion is more than welcome
Have a nice day to all
i am running it on the IOT2040 by siemens, i will try to perform an upgrade to see if this is solving the issue, but do you know any procedure i should follow to tackle the issue? i thought that throught the logging with the trace option i should see why this is happenning, but there is nothing there...
i am running it on boot using shel script, at /etc/init.d/. I will look into the syslog (thank for that),
right now i am updating it to the latest node-red (i hope) to check if this solves it
that is an interesting question, as i know they support the
but they are saying that we could upgrade by building the image ourself, i have just download the latest image from siemens (released last week), to try out, it has
NodeJS v6.14.3
Be aware that you may find nodes that will not work on an old version of NR. If you want to be current it sounds like you will have to build the image yourself
You are right, besides that i need to build my image myself, i am just seeing that NR 0.20 still work with my old node js except one node that needs version >=8.
I will try the latest image from simens and start from there.
You should be able to install the current LTS version of Node.JS by following the standard install instructions from the Node.js website. Is the CIP version of Linux based on Debian? If so, you should be able to add the appropriate repo. You might even be able to run Dave's excellent Pi script - though you may wish to ensure you can restore your device before you try.
we managed to find the cause that crashes node-red (after the suggestion to look system logs, thanks for that). We are using the node-red-contrib-rdkafka node to publish data to a kafka cluster. We see in the logs that
Blockquote user.info kernel: rdk:broker-1[437]: segfault at b2aeab9b ip b28f0b80 sp b0 2ff01c error ffff0004 in librdkafka.so.1[b28e9000+10b000]
We post the issue, but do you have any suggestion on this issue and how we could overcome it??